Being polite while fulfilling different discourse functions in online classroom discussions

Abstract

Using a discourse analytic qualitative approach, we investigated the naturally-occurring discourse that arose as part of two kinds of regular course activities, synchronous and asynchronous computer-mediated discussions. The messages contributed by members of a graduate course were analyzed for the kind of discourse functions and the kind of politeness strategies they displayed. Results indicated that synchronous CMD afforded more information seeking, information providing, and social comments than asynchronous CMD. Asynchronous discussions were slightly more likely to allow for such functions as discussion generating, experience sharing, idea explanation, and self-evaluation functions than synchronous discussions. Proportionately the two modes were similar in how politeness was expressed. Finally, in relating politeness and function, we found more politeness indicators when students were posting messages with such functions as positive evaluation and group conversation management, functions that carried the potential for face threat, and the least politeness associated with messages serving the function of experience sharing.

Introduction

As Wells (2001) wrote, “knowing is largely carried out through discourse” (p. 184), and individuals create much of the fabric of their intellectual and social lives through the words they use. Our goal in this study was to understand better how students do their discourse work when engaged in computer-mediated discussion (CMD) as part of a course activity. Instead of exploring student intent and their social concerns in CMD through interviews, self-reports, or learning outcomes, we were interested in what we could learn from an analysis of the messages themselves, the discourse produced in classroom discussion carried out online.

Two constructs were particularly relevant to this work: (a) the kinds of discourse moves or functions served by messages, and (b) the ways that messages reflect social concerns that are related to face-saving and face-threat acts, what are often called politeness strategies. We were interested in how graduate students managed politeness concerns as they fulfilled different discourse functions in CMD. In addition, we focused on studying this relationship in online tasks with different parameters, such as represented and asynchronous and synchronous discussions.

This study addressed three research questions. First, we were interested in how the messages posted by members of a graduate-level course in asynchronous and synchronous CMD represented different discourse functions. Second, we wanted to see how students used different politeness moves across the two modes of CMD, asynchronous and synchronous. Finally, we were interested in the relationship between discourse functions and the use of politeness strategies in the two CMD modes.

Increasingly prevalent in educational settings, computer-mediated discussion (CMD) allows learners to participate in and build a learning community to construct knowledge online collaboratively (Schallert and Reed, 2003–2004, Wade and Fauske, 2004). Because the discourse produced in fulfillment of course activities by learners participating in CMD is written, transcripts of online discussions can be saved easily, allowing educators and researchers to examine the discourse in order to understand students’ learning processes. As the quote from Wells (2001) above continues, “We should not be looking for learning in the time between the input from the teacher or text and later output in answers to spoken or written questions. Rather, we should expect to find the learning occurring in and through participation in the activities that make up the curriculum” (p. 184). In analyzing CMD discourse, we can examine the learning process itself rather than learning products or outcomes. As Schallert et al. (2004) reported, students give evidence of their learning even as they are involved in online discussions.

For the current study, we investigated how students learned in CMD by focusing on the functions served by different discourse moves. Our starting point was Gee’s (1996) conception of Discourse (with a capital D) as “a socially accepted association among ways of using language, other symbolic expressions, and ‘artifacts,’ of thinking, feeling, believing, valuing, and acting that can be used to identify oneself as a member of a socially meaningful group or ‘social network,’ or to signal (that one is playing) a socially meaningful ‘role’” (p. 131). In this view, Discourse is more than just language and is highly context-related. Discourse includes what happens when people draw on the contextual knowledge they have about language and on situated knowledge based on their memories of things they have said, heard, seen, or written before, in order to do things in the world (e.g., exchanging information, expressing feelings, making things happen, creating beauty, entertaining themselves and others, and so on). Discourse is both the source of this context-related knowledge and the result of it as delineated by Johnstone (2008) in the following six propositions: (1) Discourse shapes and is shaped by the world; (2) it shapes and is shaped by language; (3) the same holds for the mutual influence of Discourse and participants, (4) medium, and (5) purpose; finally, (6) Discourse is shaped by prior discourse and shapes the possibilities for future Discourse. Taking Gee’s definition of Discourse and Johnstone’s six propositions, Cameron (2001) suggested that discourse analysis should not simply focus on the linguistic forms of Discourse but also on their communicative functions. For example, the same linguistic form (e.g., a sentence with interrogative syntax) can have different functions in discourse (e.g., question or request), and the same function (e.g., request) can be realized by different linguistic forms (e.g., interrogative or imperative). Therefore, a discourse analyst should take linguistic form, context, and communicative function into consideration when analyzing Discourse. Inspired by these authors, we sought in our study to examine the discourse functions served by messages students posted in the context of a computer-mediated learning environment.

The term discourse function used here refers to the purpose of an interlocutory move occurring in an interactional event. We were influenced by Zhu’s (1996) categorization of the online asynchronous written messages of the students in a graduate seminar in terms of the following categories: information seeking question, discussing question, answer, information sharing, discussion, comment, reflection, and scaffolding. In addition, we saw the concept of discourse function as similar to the terms discourse strategies (Wade & Fauske, 2004), speech genre (Kress, 1989, Na, 2004), and social and cognitive presence (Rovai and Jordan, 2004, Rourke et al., 1999). Wade and Fauske (2004) defined discourse strategies as the ways interlocutors fulfill different goals when participating in a dialogue. Examples of goals include the desire to bring everyone to a consensus and to examine multiple possibilities and alternative hypothesis. In their analysis of transcripts from classroom CMD, Wade and Fauske identified the following discourse strategies: supporting, perspective taking, inquiring, self-questioning, challenging, nonsupporting (as in mocking or putting another person down), and posturing. Speech genre refers to types of utterances such as narrating memories and experiences, abstracting, interpreting, evaluating, eliciting, testing, etc., which are fused together to develop a speech activity (Kress, 1989). Na (2004) used the construct of genre to describe the asynchronous messages used by students in an online discussion. Finally, social and cognitive presence are two of the three constructs (along with teaching presence) that Rovai and Jordan, 2004, Rourke et al., 1999 proposed as making up what they called a Community of Inquiry. In their analysis, presence was identified from several indicators found in participants’ contributions to online discussions, with social presence referring to the degree to which participants could be socially and emotionally responsive to one another and cognitive presence referring to the various ways that learners could construct new understanding from communication. In the tradition of these researchers, our own project was an exploration of whether discourse functions (our choice of a label) could offer a framework in which messages in online discussions could be said to express the intellectual (or cognitive) and social work students were accomplishing.

In addition to discourse functions, we were interested in the issue of politeness in CMD. Yang et al. (2006) emphasized the importance of taking politeness into consideration in CMD by stating that “a concern with politeness in discourse is more than simply an additional veneer added to make one’s words ‘nicer’ but instead, seems to be at the core of reflecting how words enact or reflect the relationship between interlocutors in any discourse event” (p. 342). With his construct of online disinhibition, Suler (2004) emphasized that people can behave very differently on the Internet when compared to how they behave in face-to-face interactions with others. The emergence of the word Netiquette indicates the importance of taking courtesy into consideration when interacting in cyberspace. Combining the terms network and etiquette, Netiquette refers to a set of core rules indicating what should or should not be done in online communication to ensure common courtesy (Shea, 1994). Schallert, Cheng, and the D-Team (2008) described netiquette as the use of conventions that indicate proper manners in online communication. These core rules or conventions address the dynamic social relationships among individuals, playing an important role in the success of online learning because they affect not only cognitive but also socio-emotional processes taking place during learning (Schallert et al., 2008, Vinagre, 2008). Both the idea of Netiquette and results showing the online disinhibition effect support the necessity of studying politeness in CMD.

The concept of politeness in interpersonal interactions has a long history in the field of sociolinguisitics. With his construct of “face”, sociologist Goffman (1967) foregrounded the influential formulation of politeness that Brown and Levinson (1987) proposed. According to Goffman (1967), the term face is associated with the social-emotional notions of being embarrassed or humiliated. There is always the possibility that face can be lost, maintained, or enhanced when people participate in any interpersonal interaction. Generally speaking, when people interact with others, they not only pay attention to saving their own face, but also attend to helping others not lose face because face is mutually vulnerable. Because Goffman emphasized the universality of face concerns, Brown and Levinson (1987) assumed face as wants that all competent adult members of a society have and know each other to have.

Drawing on Goffman, 1967, Brown and Levinson, 1987, Morand and Ocker, 2003 defined politeness in CMD as an attempt to “phrase things so as to show respect and esteem for the face of others throughout social interchange” (p. 1). Face can be divided into positive and negative face. Positive face addresses the reader/hearer’s desire to be needed. Negative face addresses an individual’s desire for freedom from impingement. As one interacts in CMD, face can be threatened by acts such as disagreements, criticisms, requests for information or help, and requests for clarification of a prior message. In order to redress face-threatening acts, individuals adopt various politeness moves. Brown and Levinson (1987) categorized politeness moves into positive and negative politeness strategies, depending on which aspect of face the speaker/writer wants to save. Positive politeness strategies refer to moves “showing an appreciation of something that the speaker believes the listener would like to hear,” whereas negative politeness strategies refer to moves “attempting to reduce any imposition on the hearer” (Yang et al., 2006, pp. 341–342). Note that as Yang et al. (2006) highlighted, positive politeness and negative politeness do not refer to opposite ends of a single dimension, like a “good manners” continuum. Instead, these two categories of strategies represent the kinds of face needs they address.

Yang et al. found that the use of politeness strategies in CMD can foster a sense of community among participants by creating a comfort zone in which to exchange ideas as well as motivating students’ participation in the learning process. Yet, these same authors reported that the students in their study who were interacting online as part of a course activity sometimes showed evidence that their concerns about politeness interfered with their learning. In addition to Yang et al., other researchers have examined the politeness issue in online computer-mediated learning environments. Schallert et al. (2008) included graduate students’ self-perceptions of their own and others’ politeness to investigate whether students’ self-perceptions about their politeness concerns would be associated with their use of politeness strategies in terms of amount and kind of actual politeness moves in their online contributions. Schallert et al. used students’ self-reflection papers to understand students’ self-awareness of their politeness concerns and chose two focal students who explicitly stated that they were less concerned with politeness and three focal students who reported that they were highly concerned with issues related to politeness. They conducted a micro-discourse analysis of the written messages composed by these five selected focal students in three synchronous and three asynchronous online discussions. Results showed that the two students who were less concerned with politeness used fewer politeness moves and the politeness strategies they used had less variety; whereas the three students who self-reported as having high concerns about politeness used more politeness moves in their online discussion messages, and a greater variety was found in the kinds of politeness strategies they used.

Believing that social interaction plays an important role in the success of computer-mediated collaborative learning, Vinagre (2008) examined language learners’ use of politeness strategies in e-mail exchanges and reported that politeness influenced the efficiency and effectiveness of social interaction. In the context of e-mail tandem exchanges among college level English learners and Spanish learners, Vinagre hypothesized that students may use more negative politeness strategies than positive politeness strategies. However, findings showed that the language-learning partners used negative politeness strategies infrequently. Instead, they mainly relied on positive politeness strategies, especially those relating to claiming common ground, assuming or asserting reciprocity, and conveying cooperation. This study emphasized the different face desires addressed by positive and negative politeness strategies. Positive politeness strategies focus on closeness, solidarity, and cohesion; whereas negative politeness strategies center on formality and impersonality. Given that the goal of the messages was to allow the partners to introduce themselves to each other, it is perhaps not surprising that students attempted to establish closeness and reduce social distance rather then reflect formality between peers. The social distance between the interlocutors in Vinagre’s sample seemed to influence their preferred type of politeness strategies.

Even though Brown and Levinson’s (1987) politeness theory and most of the work on politeness has focused on human-human interaction, a few researchers have made interesting application of politeness theory to guide the sort of built-in messages computer programs would display when human users take different actions. In this use, computer systems are treated as social actors. Studies by Mayer et al., 2006, Wang et al., 2008 showed that Brown and Levinson’s (1987) politeness theory need not be limited in its application to human–human interactions, even when these interactions are mediated by the computer as in CMD, but can usefully describe how learners and users respond to and learn from different educational computer software.

Having reviewed some of the literature that has addressed discourse functions and the use of politeness strategies separately, we were interested for our study in seeing whether we could establish a systematic association between different discourse functions and different politeness moves. In this we were guided by Morand and Ocker’s (2003) suggestion that the interlocutory moves serving some discourse functions, such as presenting an alternative perspective, evaluating others’ comments, and managing the conversation, are likely to be charged with face threat. Thus, the first goal of the present study was to examine the relation between discourse functions and the use of politeness strategies in students’ postings.

Our second goal was to explore whether the relationship between discourse function and politeness strategies evident in one mode of online discussion would be represented similarly in another mode. Because the setting to which we had access was a course that incorporated several modes of communication including synchronous and asynchronous discussion tool, we had a natural opportunity to test the extent of the generality of claims about the relationship between discourse function and politeness moves. We were interested in whether ways of expressing politeness and of fulfilling different discourse functions would be affected by the affordances of these communication modes (Box, 1999). To communicate through an asynchronous mode, as on message boards, participants are not required to be online at the same time. Because messages can be posted, and encountered and responded to by another student coming to the discussion at some later time, students can consider what to say, whether to say it, and how to phrase their messages more carefully than when engaged in the tumultuous exchange characteristic of synchronous discussions. The latter, by contrast, allowing as it does for real-time interaction among participants who are online simultaneously, forces students to choose between posting their contributions quickly with little review or becoming hopelessly behind in reading others’ postings and seeing what others have responded as the discussion unfolds.

In terms of the implied comparison that our study entailed, there have not been many studies directly comparing in strict experimental fashion characteristics and effects of synchronous and asynchronous modes of discussion. Instead, studies typically include only one or the other mode. For example, studies of the discourse of asynchronous CMD have focused on several different aspects including knowledge construction (e.g., Gunawardena et al., 1997, Henri, 1992, Zhu, 1996); social presence (e.g., Rourke et al., 1999); interaction patterns (e.g., Fahy et al., 2000); learning strategies (e.g., Lockhorst, Admiraal, Pilot, & Veen, 2003); and community building (e.g., Butler, Sproull, Kiesler, & Kraut, 2008). Compared to the amount of research conducted on asynchronous CMD, studies of synchronous CMD are relatively rare (see the following as examples: Abrams, 2001, Kneser et al., 2001, Schallert et al., 1996).

Recently, some researchers have attempted to compare learning occurring in asynchronous and synchronous CMD mode (e.g., Abrams, 2003, Chou, 2001, Johnson and Johnson, 2006, Perez, 2003, Thomas and Macgregor, 2005). For example, Abrams (2003) compared three groups of students learning German in terms of their subsequent oral performance, one group that had participated in synchronous online discussion, one that had participated in asynchronous discussion, and a control group. Only the synchronous group differed from the other two, showing a significant increase in quantity of oral language produced subsequent to online synchronous discussion. Chou (2001) reported that during the online discussion, students were more likely to engage in task-oriented rather than in social and off-task messages whether in the synchronous or asynchronous online activity, although the synchronous mode led to a higher proportion of social and emotional messages than the asynchronous. When contributing to the asynchronous discussion, students seemed more interested in presenting their own opinions whereas they seemed more interactive and interested in the views of their peers in the synchronous discussion mode. Thomas and Macgregor (2005) reported on the online activities of undergraduate students involved in a problem-based assignment, finding that the asynchronous mode was preferable for tasks that required deep reflection whereas the synchronous mode was best for aspects of tasks that needed brainstorming and building group cohesion.

In this project, we were interested in the discourse moves of students as they interacted with each other as part of a classroom activity. Our focus was on the kind of face “work” the messages revealed by way of politeness strategies as students attempted to accomplish different moves or functions with their postings. Because of the different affordances made possible in the two modes of discussion (synchronous and asynchronous) that were a part of the course we were studying, we were interested in whether the relation between particular politeness strategies and discourse functions would hold similarly across the two modes of online discussion. Our research addressed the following questions:

• What discourse functions are evident in the postings of students participating in synchronous and asynchronous discussions?

• What kinds of politeness strategies do students use as they participate in synchronous and asynchronous classroom discussions?

• What is the relationship of discourse function to politeness strategy in the two kinds of online discussion, synchronous and asynchronous?